Design Of Graphic Reactive Power Optimization Software And Research On Algorithms Of Reactive Power Optimization

In practical application, reactive power optimization (RPO) in power systems is one of the most effective control methods to ensure power system operation securely and economically, which is an important measure to improve the voltage quality and reduce the net real loss. In theory, RPO is a non-convex and multimodal complex optimization problem involving nonlinear objective function, nonlinear equality and inequality constraints and both continuous and discrete variables. It is quite difficult to be solved quickly and accurately. In this thesis, the design of the practical graphic reactive power optimization software and the algorithms of reactive power optimization are studied thoroughly.The main contributions of the thesis are as follows.Based on the practical background of Tongliao transformer station reactive power optimization project, the hardcore of this system, objects of equipment's management based on object persistence technology and technology of multi-documents has been studied. Moreover, the design and achieve of objects of equipment using the serialized class of Microsoft Foundation Classes (MFC), and complex link date structure, is discussed intensively.Mathematical models of reactive power optimization problem and the algorithm of reactive power optimization are studied thoroughly. In this thesis, with the shunt capacitance /reactance parameter and the ratio of transformer parameter, a new mathematical model of reactive power optimization problem is deduced in details. With this model, the relationships between variables of reactive power optimization problem are clearer.To solve the complex optimization problem involving constraints, an algorithm consisting of "multiplier method", "BFGS quasi-Newton method" and "inaccurate search for step" is designed. It is applied to deal with the reactive power optimization problem.On the basis of careful research on genetic algorithm, particle swarm algorithm and artificial fish swarm algorithm, an integration optimization program is achieved. In this thesis, with the characteristics and framework of intelligent algorithms, the immune information processing mechanism of immune system is involved into original artificial fish swarm optimizer. The proposed algorithm can improve the abilities of seeking the global excellent result.The methods to deal with discrete variables in the complex optimization problems are also discussed detailedly. Using real-integer coding of genetic algorithms to deal with the discrete variables and continuous variables directly; pushing discrete variables'values to the nearest integer values during the iterative process in particle swarm algorithm; using a quadratic penalty function to drive the discrete variables toward their neighborhood centers during the iterative process in artificial fish swarm algorithm.Furthermore, the optimization algorithms proposed in the thesis are programmed, and the computing results against the IEEE 30-bus system show the effetiveness of the proposed algorithms of reactive power optimization.